Research map / Detection and segmentation / CNN features
Context-Reinforced Semantic Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add RL-selected context from predicted segmentation maps | Baseline segmentation network · Cityscapes validation | mIoU: 76.36 -> 79.10 | Helped |
| Add RL-selected context from predicted segmentation maps | Baseline segmentation network · ADE20K validation | mIoU: 40.97 -> 42.56 | Helped |
| Use the complete p-map directly as context instead of learning which information to select (ablation) | Baseline segmentation network · Cityscapes validation | mIoU: 76.36 -> 75.19 | Hurt |
| Use the complete p-map directly as context instead of learning which information to select (ablation) | Baseline segmentation network · ADE20K validation | mIoU: 40.97 -> 39.44 | Hurt |
| Replace RL context learning with gate-like selection and strongly suppress exploration (ablation) | Baseline segmentation network · Cityscapes validation | mIoU: 76.36 -> 77.23 | Helped |
| Replace RL context learning with gate-like selection and strongly suppress exploration (ablation) | Baseline segmentation network · ADE20K validation | mIoU: 40.97 -> 41.31 | Helped |
| Add context by using 2 iterations rather than no context (ablation) | CiSS-Net · Cityscapes test | mIoU: 77.59 -> 79.21 | Helped |
| Add context by using 2 iterations rather than no context (ablation) | CiSS-Net · ADE20K validation | mIoU: 40.97 -> 42.56 | Helped |
| Increase context iterations from 2 to 4 (ablation) | CiSS-Net · Cityscapes test | mIoU: 79.21 -> 79.29 | Helped |
| Increase context iterations from 2 to 4 (ablation) | CiSS-Net · ADE20K validation | mIoU: 42.56 -> 42.63 | Helped |
| Increase context iterations from 4 to 6 (ablation) | CiSS-Net · Cityscapes test | mIoU: 79.29 -> 79.29 | No clear effect |
| Increase context iterations from 4 to 6 (ablation) | CiSS-Net · ADE20K validation | mIoU: 42.63 -> 42.66 | Helped |
Closest papers by meaning
- Context Encoding for Semantic Segmentation (CVPR 2018 · similarity 0.89)
- Context Prior for Scene Segmentation (CVPR 2020 · similarity 0.89)
- Contextrast: Contextual Contrastive Learning for Semantic Segmentation (CVPR 2024 · similarity 0.89)
- Adaptive Pyramid Context Network for Semantic Segmentation (CVPR 2019 · similarity 0.89)
- Learning Context-Aware Classifier for Semantic Segmentation (AAAI 2023 · similarity 0.88)
- Reinforced active learning for image segmentation (ICLR 2020 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.